Prevention of the angiopathic consequences of long-term human diabetes remains the primary goal of diabetes therapy. One possible approach to this problem is the development of a totally implantable, artificial "beta cell", to provide "physiologic" glucose homeostasis in the diabetic patient. The main goals of the proposed research were formulated in anticipation of the development of such a device, and in view of the progress being made in several centers toward the development of an implantable glucose sensor. The primary aim of the present study is the modification and testing of our implantable infusion pump to include a flow regulating capability; the entire package would then be suitable for inclusion into a artificial beta cell. The first phase of the project was to assess and solve the problems of insulin's heat lability, binding to inert pump components, and possible effects on local blood coagulation during chronic in vivo insulin infusion. The next phase will involve the design and in vitro testing of two types of flow regulators, to provide both continuously variable flow and two rates of constant infusion, one of the rates being controlled by an on-off valve. The final project phase will be the long-term in vivo testing of pumps plus flow regulators in experimentally diabetic animals.